Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIDeepSeekDeepSeekMuseMuse+22 more

    Spec-Driven Development Toolkit

    1

    Turn feature requests into reviewable specs and traceable implementation tasks—with approval gates, brownfield support, and built-in Python helpers.

    $9.99

    Secure checkout via Stripe

    Money-back guarantee

    Converts to your local currency at checkout

    Security scanned
    Spec-Driven Development Toolkit

    Spec-Driven Development Toolkit

    Example session with this skill installed

    Add a “Copy link” action to the project share page. Draft a compact requirements document covering successful copy, clipboard failure, and accessible confirmation. Use existing project context. Stop after requirements and ask for approval; do not create a design or modify code.

    • Read your context and instructions
    • Compiled the spec-driven development toolkit

    Requirements: Copy Share Link

    Status: Draft

    Goal

    Let users copy the current project share URL from the share page.

    Requirement 1: Copy the share URL

    User Story: As a user viewing the share page, I want to copy its URL so I can share it with others.

    Acceptance Criteria

    1. WHEN a user activates “Copy link” and clipboard access succeeds, THE App SHALL copy the current share URL and show a confirmation.
    2. IF clipboard access fails, THEN THE App SHALL present the URL in a selectable form and explain that it must be copied manually.

    Requirement 2: Accessible confirmation

    Acceptance Criteria

    1. WHEN the copy action succeeds, THE App SHALL make the confirmation available to assistive technology.

    Out of Scope

    • Changing who can access the shared project.
    • Creating or expiring share links.

    Does this match the intended behavior? Please approve or give feedback. I will not create a design or modify application code until you approve.

    Connects securely to your tools. The creator never sees your data.

    About this skill

    Plan before coding—and keep implementation tied to decisions

    Spec-Driven Development gives AI coding agents a repeatable workflow for turning feature requests into reviewable specifications and traceable implementation work. It uses EARS (Easy Approach to Requirements Syntax), a lightweight pattern-based method for writing clear, testable requirements in natural language.

    What it creates

    • requirements.md: a glossary, user stories, and numbered acceptance criteria using EARS patterns.
    • design.md: architecture, component interfaces, data models, error handling, test strategy, and a requirements traceability table.
    • tasks.md: an incremental implementation checklist whose tasks reference requirement IDs.
    • bugfix.md: current, expected, and unchanged behavior for defect-focused work.

    Six workflows

    1. Create project steering documents from verified repository evidence.
    2. Draft a feature spec in three approval-gated phases: requirements, design, and tasks.
    3. Specify bugs while protecting adjacent behavior from regressions.
    4. Execute one approved implementation task at a time and verify it before marking it complete.
    5. Propagate change requests through the affected spec documents before changing code.
    6. Adopt SDD incrementally in an existing codebase by documenting observed behavior instead of guessing intent.

    Included

    • Reusable Markdown templates for specs and project context.
    • Python 3 scripts to scaffold specs and check requirement-to-task traceability; no third-party packages are required.
    • Guidance for EARS criteria, project steering, and spec change workflows.

    The skill writes prose in the language you use. During planning, it pauses for approval at each phase; implementation begins only when you request task execution. Testing and human review remain necessary—the workflow does not guarantee correctness or perform deployment or manual QA.

    Requires an AI agent that supports SKILL.md. Python 3 is needed only for the optional helper scripts. The purchase includes a perpetual license for use in the purchaser's own projects; the skill itself may not be redistributed. See the included LICENSE for terms.

    Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Prompt & Skill Engineering.

    Requires an AI assistant that supports the SKILL.md standard. Python 3 is required only to run the included helper scripts. The workflow depends on accurate project context and user review; it does not guarantee defect-free code, replace testing or manual QA, or perform deployment.

    How to use

    1. 1

      Connect your agent to the Agensi MCP

      Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup

    2. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 3

      Ask for what you need

      Your agent loads this skill and uses it on your task. It always asks before buying a skill.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill goes through the trust checks below.

    Security scanned

    Checked by our 8-point scan before listing

    Fresh listing

    Recently published to Agensi

    Money-back guarantee

    Doesn't do what its page promises? You get your money back.

    Trust & safety

    Security scan

    Security scan passed 10 days ago

    Nothing to install. It runs in your AI agent.

    Optional
    Python 3

    Python only runs the included helper scripts.

    Permissions

    Terminal / Shell
    Read Files
    Write Files

    File Scopes

    **/*

    Reads relevant project context and code. Writes spec files and, only when asked to execute approved tasks, implementation files required by those tasks. No browser, network, account credentials, or environment-variable access is needed.

    Listed10 days ago

    What's inside

    Frequently Asked Questions

    A self-contained skill package with SKILL.md, reusable templates, reference guides, and two Python helper scripts for scaffolding specs and checking requirement-to-task traceability. The purchase includes a perpetual license for use in the purchaser's own projects; the skill itself may not be redistributed. See the included LICENSE for the exact terms.

    No. For a new spec, the agent drafts requirements, design, and tasks with your approval at each phase. It stops before implementation; code changes begin only when you request task execution. The default is one approved task at a time.

    Yes. It adopts an existing project incrementally: the agent records current behavior from repository evidence, notes open questions instead of guessing intent, and creates a baseline for the area you plan to change.

    New in Prompt & Skill Engineering